Juju 2.5 fails to deploy canonical-kubernetes charm in AWS

[Kenneth Koski]

I ran these two commands with Juju 2.5-rc1-bionic-amd64:

juju bootstrap aws/us-east-1
juju deploy cs:bundle/canonical-kubernetes

The bundle did not boot up successfully. Here is the output from `juju status`:

Model Controller Cloud/Region Version SLA Timestamp
default aws-us-east-1 aws/us-east-1 2.5-rc1 unsupported 14:43:20-06:00

App Version Status Scale Charm Store Rev OS Notes
easyrsa maintenance 1 easyrsa jujucharms 195 ubuntu
etcd waiting 1/3 etcd jujucharms 338 ubuntu
flannel waiting 0 flannel jujucharms 351 ubuntu
kubeapi-load-balancer waiting 0/1 kubeapi-load-balancer jujucharms 525 ubuntu exposed
kubernetes-master waiting 0/2 kubernetes-master jujucharms 542 ubuntu
kubernetes-worker waiting 0/3 kubernetes-worker jujucharms 398 ubuntu exposed

Unit Workload Agent Machine Public address Ports Message
easyrsa/0* maintenance executing 0 3.84.147.43 (install) installing charm software
etcd/0* maintenance executing 1 54.236.234.58 (install) installing charm software
etcd/1 waiting allocating 2 100.27.47.226 waiting for machine
etcd/2 waiting allocating 3 34.226.153.176 installing agent
kubeapi-load-balancer/0 waiting allocating 4 52.90.158.171 waiting for machine
kubernetes-master/0 waiting allocating 5 35.172.232.105 installing agent
kubernetes-master/1 waiting allocating 6 54.172.192.101 waiting for machine
kubernetes-worker/0 waiting allocating 7 waiting for machine
kubernetes-worker/1 waiting allocating 8 waiting for machine
kubernetes-worker/2 waiting allocating 9 waiting for machine

Machine State DNS Inst id Series AZ Message
0 started 3.84.147.43 i-0bca6b9cef81a1068 bionic us-east-1a running
1 started 54.236.234.58 i-03300106c795e5fcd bionic us-east-1a running
2 pending 100.27.47.226 i-02cd7f99a20d3052f bionic us-east-1b running
3 started 34.226.153.176 i-0a9443ddb7c2972ee bionic us-east-1c running
4 pending 52.90.158.171 i-0cd3f8fe4c8c625f8 bionic us-east-1d running
5 started 35.172.232.105 i-001909bd53a164f7e bionic us-east-1b running
6 pending 54.172.192.101 i-0f89ec869d663ef61 bionic us-east-1a running
7 down pending bionic cannot run instances: The requested configuration is currently not supported. Please check the documentation for supported configurations. (Unsupported)
8 down pending bionic cannot run instances: The requested configuration is currently not supported. Please check the documentation for supported configurations. (Unsupported)
9 down pending bionic cannot run instances: The requested configuration is currently not supported. Please check the documentation for supported configurations. (Unsupported)

`juju show-machine 7` gives this output:

model: default
machines:
  "7":
    juju-status:
      current: down
      message: agent is not communicating with the server
      since: 10 Jan 2019 14:43:07-06:00
    instance-id: pending
    machine-status:
      current: provisioning error
      message: 'cannot run instances: The requested configuration is currently not
        supported. Please check the documentation for supported configurations. (Unsupported)'
      since: 10 Jan 2019 14:43:07-06:00
    series: bionic
    constraints: cores=4 mem=4096M root-disk=16384M

If it helps at all, the instance type it was attempting to spin up was "a1.xlarge".

Deploying the same bundle with Juju 2.4 works fine.

[Ian Booth (wallyworld)]

https://github.com/juju/juju/pull/9622

Juju was using some incorrect aws instance data that didn't account properly for arm64 instances.